Re: The cost of stack traces

[Bryce McKinlay <mckinlay at redhat dot com>]

Andrew Haley wrote:
> One other thing that would solve my problem entirely: if addr2line
> fails to return any information about a symbol, blacklist that entire
> object file.  This might give suprious results if, e.g., only part of
> the object file had debbuginfo.  However, it would remove the
> thousands of fruitless searches we have at the present time.
>   

I've checked in a patch to implement this: 
http://gcc.gnu.org/ml/java-patches/2006-q2/msg00137.html

I think its pretty safe for now to assume that a binary either has 
.debug_line info or it doesn't.

Bryce